The Nintendo Switch 2 hybrid console is due to launch globally next Wednesday (June 4). The highly anticipated next-gen handheld is powered by a custom NVIDIA processor. To the surprise of many industry watchdogs, both parties have semi-recently disclosed a couple of technical details regarding their fruitful hardware collaboration. Historically, Nintendo has guarded many aspects of its past generation hardware. Throughout the 2020s, data miners and leakers have unearthed plenty of pre-release information—leading to theories about the Switch 2 chipset's origins. During the Switch 1 era, TSMC was the chosen manufacturing partner. NVIDIA's off-the-shelf Tegra X1 mobile SoC powered the first wave of Nintendo Switch (2017) devices, in 20 nm form. A 2019 revision resulted in Switch Lite and (refreshed) Switch models being equipped with a more efficient 16 nm solution, also present within 2023's premium OLED variant.

Since then, Switch 2's alleged NVIDIA Tegra T239 SoC was linked to a Samsung 8 nm node process. Earlier this month, extremely brave Chinese leakers produced "full die shot" evidence of South Korean foundry origins. Bloomberg insider news articles have implied that Samsung Semi's mature 8 nm FinFET node is better suited—rather than an equivalent TSMC product—for the Switch 2's custom NVIDIA chipset. Unnamed sources have mentioned critical factors; namely stable production and process compatibility. Industry moles reckon that Samsung leadership is actively and aggressively pushing for a longer Switch 2 chipset production deal. Renewed terms could include a future die shrink; pre-launch analysis indicates a sizeable 207 mm² footprint. Beyond foundry biz negotiations, additional murmurs suggest company executives dangling an OLED panel supply agreement. Industry experts have viewed Samsung's key entry—into the gaming console chip market—as a seismic development. A DigiTimes article dives into a so-called "tripartite cooperation"—involving Nintendo, NVIDIA, and the South Korean semiconductor giant. The Samsung Foundry has floundered and struggled in recent times, but is keen to catch up with its arch rival. Fresh rumors have AMD and Sony considering Samsung's chip making channels; possibly with futuristic PlayStation hardware in mind.

A fair few partner-produced items—exhibited at last week's Computex 2025 trade event—were blessed with Jensen Huang's signature. Not long after concluding his keynote presentation, the ever intrepid NVIDIA boss spent time on the showroom floor. The TechPowerUp crew stumbled upon an autographed example at the Palit booth: Pandora NXNano mini AI PC. Due to mitigating circumstances, we did not capture any shots of a nearby special prototype flagship graphics card. The Taiwanese manufacturer proudly showcased a "GeForce RTX 5090 Tornado" dual-fan/dual-slot design; its all-metal housing aesthetic recalls an older vintage of Team Green's Founder Edition line.

Several media outlets have examined this compelling prospect; at least from a safe distance. Palit did not allow any hands-on experiences. A first party-developed "double flow-through cooling solution" serves as a main inspiration for the Tornado engineering team. The GeForce board partner's implementation of an angled 12v-2x6 power connector is another borrowed element. According to VideoCardz's expert opinion, NVIDIA has supplied a reference PCB. Palit is best known for making budget-friendly (mostly plastic) options, so their Tornado card concept represents a compelling breakaway from normal patterns. Despite gaining an important official endorsement; several rival products were similarly decorated with Huang's autograph.

Earlier this month, Bryan Bilowol—of Tech YES City fame—was taken aback by the unexpected "death" of his Ryzen 9 9950X processor, at the alleged hands of an ASRock X870 Steel Legend motherboard. Throughout early 2025, widespread news reports and community tracking efforts have largely implicated ASRock boards. AMD's highly popular Ryzen 7 9800X3D gaming CPU has "fallen victim" on numerous occasions, but a smaller number of incidents have affected non-3D V-Cache equipped Ryzen 9000 "Granite Ridge" models. The Taiwanese manufacturer and Team Red have investigated these problems, and have released fixes via BIOS updates. Despite these efforts, unhappy owners have experienced CPU failures in Q2'25. Following the uploading of initial findings and theories, Bilowol promised further reports. Fortunately, a sit-down—with ASRock's motherboard team—happened late last week, at Computex 2025. At the time of writing, the company has not published an official statement regarding their latest findings or fixes. Tech YES City shared a couple of compelling details in a May 26 YouTube upload (viewable below).

Commenting on the Ryzen 9000 CPU failure situation, Bilowol relayed important ASRock material: (the motherboard team) told me it had to do with the EDC and TDC, which is the electric design current as well as the thermal design current. Essentially, they're saying it's an ampage problem that exists with the precision boost overdrive (PBO) settings. In particular, these mid-range and high-end motherboards—say for instance a B650E or an X670E Taichi or even a B850 Steel Legend—have the ability to run Ryzen 9000 series CPUs at their maximum PBO settings out of the box. What's happening here is ASRock told me they were implementing these settings to what they believe was too aggressive for what the CPUs were able to handle, at least for earlier samples. However, they do promise me that these issues have been fixed with their latest BIOS updates that they're rolling out for these mid-range and high-end motherboards. They also told me that the lower-end boards, say for instance, an A620 HDV or something like that, shouldn't be affected by this issue simply because those PBO settings have been deliberately tuned down, due to the motherboard in question not being a high-end model." Three months back, ASRock's Japanese branch accused the PC hardware community of spreading "misinformation." Hopefully, the latest batch of fixes will bring an end to all Ryzen 9000 CPU + high-end/mid-range ASRock mainboard controversies.

Prior to AMD's official unveiling of the Radeon RX 9060 XT series, Sapphire's Computex booth teased a forthcoming lineup of custom graphics card options. Hours later, the TechPowerUp crew inspected freshly wall-mounted specimens. To the surprise of many, the Hong Kong-based manufacturer is readying a premium Radeon RX 9060 XT 16 GB NITRO+ SKU for launch time (on June 5). The usual suspects—in PULSE and PURE guises—were also exhibited within the walls of the Taipei Nangang Exhibition Centre. Not long after the closing of ceremonies, Sapphire China's JD.com store has opened up pre-orders for six launch day products.

Yesterday, a VideoCardz investigative piece delved into pre-release price points (including VAT). Keen potential customers can secure finalized retail units via a deposit system; demanding a 50 RMB (~$7 USD) upfront fee. Unsurprisingly, the largest tag is affixed to Sapphire's triple-fan 16 GB NITRO+ model—3299 RMB (~$458 USD). This top-tier option sits 800 RMB above Team Red's suggested guideline. The AIB's barebones dual-fan Radeon RX 9060 XT PULSE 8 GB package adheres to official MSRP: 2499 RMB (~$347 USD). Their PULSE 16 GB model sits almost in the middle of the lineup—in terms of pricing—at 2899 RMB (~$402 USD). VideoCardz believes that another PULSE color/finish variant will emerge, at some point post-launch—bringing the total number of items up to seven distinct products.

During a recent Gigabyte Nan-Ping factory tour, members of the media/press encountered trays of brand-new bare motherboards and graphics cards. HWCooling's Jan Olšan noticed the presence of a not-yet-launched RDNA 4 product. Fortunately, the manufacturer granted photo access (two examples are below). A full report outlined multiple products and production pipelines, but Olšan's scoop serves as a nice preview of things to come. We anticipate TechPowerUp's W1zzard de-shrouding of numerous custom Radeon RX 9060 XT designs, early on next month. Prior to the publication of comprehensive evaluations, HWCooling's article provided an early insight: "a wild Radeon RX 9060 XT appears…take note of the graphics card in the top-right corner of the first lot of PCBs—this is likely the first publicly shown PCB of the Radeon RX 9060 XT, which, technically speaking, may still be under NDA. So enjoy this privileged look at the exposed Navi 44 chip."

Olšan linked the factory specimen to a freshly revealed design: "the card will go on sale June 5, with reviews a day earlier. Its design matches AMD's official renderings, and its identity is further pointed at by the unusual three display connectors. This should be the PCB for the Gaming version of the card, or more specifically, the Gaming OC model with three fans. The cooler will significantly overhang the PCB, with a pass-through window under the last fan." Last week, the TechPowerUp team spent hands-on time with Gigabyte's premium Radeon RX 9060 XT Gaming OC SKU. Prior to Computex 2025, product registration information pointed to the Taiwanese manufacturer readying Gaming OC 16 GB and 8 GB variants. Curiously, the company's RDNA 4 custom card portfolio is represented by a lone ultra-premium AORUS ELITE flagship option and various triple-fan Gaming OC models.

In the run-up to Computex 2025, Intel sent out mixed messages about the future of Arc Xe2 "Battlemage" desktop graphics cards. A combination of first-party teasers and "official leaks" (courtesy of AIBs) indicated an (eventual) introduction of professional B-series products. PC hardware enthusiasts were looking forward to a possible unveiling of higher-end gaming cards; the rumor mill has floated a "B770" option in recent times. Industry observers detected further encouraging pre-release signs; a "B750" identifier was discovered on Intel Japan's website—likely existing as placeholder material. Team Blue's social media account drummed up additional hype, prior to the commencement of last week's key trade event.

The TechPowerUp crew and other media outlets did not stumble upon any Intel Arc Xe2 B750 or B770 specimens at the Taipei Nangang Exhibition Center. According to a mid-visit Tweakers.net article, several unnamed figures (likely board partner company representatives) divulged inside track knowledge during Computex 2025. The online publication outlined a smattering of details: "sources close to Intel confirm the existence of the Arc B770 video card. According to them, the introduction is planned for the second half of this year. The card is expected to appear in the fourth quarter, although the planning can of course still change. Intel is working on an Arc B770 video card based on the 'Battlemage' architecture. Tweakers was able to confirm this with multiple sources during Computex. So far, Intel has only released lower-positioned B570 and B580 video cards with this architecture." Now defunct product roadmaps had the "Battlemage" series wrapped up by 2024, but delays and other factors have allegedly caused major timeline slippages. An industry expert claimed that a larger "BMG-G31" GPU development project was closed down at some point late last year, but fresher production activities have generated renewed beliefs that the "Battlemage" gaming line will expand beyond "cheaper" tiers.

We're close to the finish line on Framework Desktop development. We're now at the end of the DVT2 phase where we complete validation and finalize firmware and drivers. With that, we're continuing our deep dive series, this time digging into the semi-custom 400 W Power Supply we integrated into the product. This may sound mundane, but it's one of the more complex parts of the product. To align to our product philosophy, we used a standard FlexATX form factor. That means both that you can bring your own compatible power supply or re-use the one we designed in the future with any other ATX-compatible system. If you're bringing your own, make sure to pick one that can handle at least 32.5 A on the 12 V rail.

Although the "peak" power demand of the AMD Ryzen AI Max processor is 140 W, it can actually pull hundreds of watts in millisecond-level bursts. Because there is no battery in the Framework Desktop to absorb this, we sized up the Power Supply to handle it and maximize performance. We worked with FSP, one of the biggest power supply makers in Taiwan and the manufacturer behind many other brands, to build a compact, high-efficiency, quiet design. We actually started with one of their higher wattage off the shelf power supplies, down-rated it to 400 W to focus on efficiency while allowing higher peak load, and customized it from there. We modified a few areas to optimize it for the core Framework Desktop use cases.

The majority of PowerColor coverage—emerging from Computex 2025—focused on custom Radeon RX 9060 XT graphics cards. The usual staple of Reaper and Hellhound options were showcased in Taiwan, but a brand-new family made an appearance. As highlighted by Shroud/syurarara (a YouTuber), a whole corner of the brand's booth was dedicated to a lone demonstration unit. Not much context was provided, beyond a logo and an accompanying mascot/icon/protagonist. Shroud shared a couple of tidbits: "apparently a new PowerColor model will be released in the future! It's a mix of white and gray. The name of the new graphics card hasn't been decided yet, but the female character's name is apparently REVA."

Given the lack of specifications and little explanation (provided by company representatives), PowerColor could jump-start a full-blown marketing campaign in the near future. Their brand-new REVA graphics card series seems to be semi-related to the familiar Hellhound IP; observers noticed the presence of two Hellhound emblem-branded cooling fans on the Computex sample/prototype. Despite having an almost direct line to a source at (or adjacent to) PowerColor, VideoCardz did not get an inside scoop on REVA. Their news report opined that the white/gray card could leverage Radeon RX 9070 XT (or non-XT) hardware, and be lined up for a possible global release.

XPG's newly unveiled SFF-ready platform—carefully optimized for NVIDIA's omnipresent GeForce RTX 50-series—consists of several product ranges. Given the complicated nature of the manufacturer's various new item lineups, this article will concentrate on freshly unveiled power supplies units. At Computex 2025, ADATA's gaming division has keenly showed off its brand-new PYMCORE SFX PSU series. The company's ultra-compact designs are tailored for integration into VALOR AIR and MESH NANO cases—this synergistic combination is advertised as: "allowing longer GPUs to fit into SFF builds without compromising performance."

The XPG PYMCORE power supply series is fully up-to-date with modern standards—most notably, being ATX 3.1 and PCIe 5.1 compliant. A 1000 W Platinum 80+ certified flagship model offers 235% power excursion. 12V2x6 (600 W) connectors can be linked to compatible SFF-ready enthusiast-grade NVIDIA GeForce graphics cards (and a smattering of custom Radeon RX 9070 Series models). A flagship PYMCORE unit was closely inspected by a TechPowerUp staff member; they took note of three 12V2x6 outputs. One exhibited specimen was deemed very special; due to having been blessed by Jensen Huang's signature. In addition, XPG is readying 850 W and 750 W Gold-rated PYMCORE SKUs. All models are configured with 100% Japanese capacitors.

Zalman has debuted its ZM-DF120 fan series at Computex 2025; with an emphasis on reducing cable clutter and easy installation. This is enabled by a magnetic mounting system and per-unit pogo pin interfaces. Similar designs have emerged from rival companies in the recent-ish past; eventually leading to ugly legal battles. The South Korean manufacturer is (bravely) stepping into this arena with its heavily ARGB-lit ZM-DF120 120 mm PWM fan options; showcased in black and white finishes. Side infinity mirror panels are hyped as providing dazzling visual effects. According to specification placards, fan speed range is 600-2000RPM (±10%). Maximum noise level is 34.5 dB (±10%), and max. air flow is 70.05 CFM (±10%). Zalman did not list any static pressure data. As expected, their ZM-DF120 fans leverage a hydro bearing platform.

Zalman has showcased their ZET series at Computex 2025; a brand-new family of air CPU tower coolers. This lineup seems to serve as a natural successor to past shell-less CNPS offerings. Circular cooling solutions are an unusual sight in the 2020s, but the South Korean manufacturer is intent on breaking beyond current (boxy) trends. Company representatives and product placards mentioned a main shape inspiration: aircraft engine designs.

The flagship ZET5 model was presented in black and white forms; both augmented with tasteful ARGB lighting rings. The brand's premium option has a 200 W TDP rating. Two modular fans are attached via a magnetic system, and connectivity is granted by a pogo pin system. The ZET5 features copper heatpipes, and a surprisingly non-radial aluminium fin array. A tube of ZM-STC10 thermal compound will be bundled in with these top-end products. Tentative pricing was not available at the time of writing.

Earlier today, the Sapphire Japan social media account shared an intriguing teaser photo. As expected, the Hong Kong-based manufacturer has (representative) feet on the ground in Taipei, Taiwan. Currently, their Computex booth's main purpose is to show off already launched custom AMD RDNA 4 products—namely Radeon RX 9070 XT and RX 9070 cards. According to Team Red's official schedule, new items will be debuted on-stage tomorrow—insiders believe that the oft-leaked Radeon RX 9060 XT model will be unveiled by Jack Huynh, and board partners.

The Japanese Sapphire office commented on pre-game conditions: "I received a very interesting image from the site. Um...It says something like "tomorrow, the 21st, from 12:01pm (1:01pm in Japan?)" I wonder what it could be." Their photo upload shows five placeholder placards; two with longer profiles—suggesting triple-fan configurations. The three other options seem to be dual-fan in nature. Printed material outlines an imminent "21 May, 12:01 PM" reveal. Despite fairly strict conditions, other AIBs have already presented designs (online). Yesterday, industry watchers noticed Acer's previewing of a Nitro Radeon RX 9060 XT OC 16 GB SKU. Given the total number of "temporary" wall-mounted items on display, Sapphire could be readying various mid-range (PURE) and budget (PULSE) models.

Cooler Master has revived its gargantuan Cosmos case product line; as demonstrated by the brand dedicating a whole table to two behemoth samples. Massive chassis designs are not particularly "de rigueur" nowadays, but 2025 revivals could be appropriate fits for ever-growing graphics card profiles, and extra elaborate cooling setups. Two different configurations were presented; as denoted by product placards and differentiated by modular panel selections. This year's refresh supports modern hidden and back-connector products.

The Cosmos 2025 "Water Cooling" demo unit's foundation consists of a non-specific Intel Ultra 5 processor, a customized (watercooled) GeForce RTX 5090 card, and an ASRock Z890 LiveMixer WIFI mainboard. Cooler Master has installed their fancy X Pro waterblock system; the visiting TechPowerUp team noticed that these components occupied only a small portion of internal volume. The 180 mm long MWE Gold 1250 V2 PSU looks relatively dinky, in its elevated perch. A radiator setup seems to feature absolutely huge airflow fans (maybe four in total); two of these units are visible in some of our photos.

Inno3D is exhibiting many already released graphics card products at their Computex 2025 booth, but TechPowerUp's day zero visit focused on an intriguing wall-mounted demo unit. Earlier today, NVIDIA announced the official launch of its affordable GeForce RTX 5060 8 GB desktop and laptop designs. Prior to full-on introduction, media outlets have expressed dissatisfaction regarding Team Green's "Blackwell" GPU entry-mainstream review conditions. Currently, many professional reviewers are "on-site" in Taiwan. At this rate, the publication of in-depth evaluations will occur following the conclusion of this week's extremely important trade event.

The Inno3D GeForce RTX 5060 TWIN X2 OC White SKU reuses a familiar dual-fan setup; as featured on readily available GeForce RTX 5060 Ti 8 GB options (black/silver or white/silver). Interestingly, the manufacturer also extends this dual-slot thick design into triple-fan territories (for RTX 5060 Ti and above). The TechPowerUp GPU database does not allude to any Inno3D GeForce RTX 5060 (non-Ti) models being outfitted with a third fan. As discussed in a mid-April PR piece, Inno3D engineers have drummed up a custom PCB layout. Despite the implementation of a very shallow overclock, the TWIN X2 cooling solution is not expected to struggle—when tasked with the tempering of Team Green's GB206 GPU die.

Acer's dedicated Computex 2025 preview landing page mostly concentrates on brand-new AI PC products, but eagle-eyed observers have noticed the inclusion of an NDA-busting item. The manufacturer's web marketing team has inadvertently unveiled their forthcoming Nitro Radeon RX 9060 XT OC 16 GB model; days in advance of AMD's official unveiling (May 21). Over a week ago, momomo_us shared a screenshot of an unnamed retailer's listing of not-yet-official Acer Nitro 16 GB and 8 GB SKUs. The keen tech industry watcher followed up with today's discovery; complete with a promotional render.

As expected, the Taiwanese brand has prepared a new dual-fan design—borrowing elements from semi-recently revealed Radeon RX 9070 Series cards. Despite rumors of leveraging lesser hardware—probably a "Navi 44 XT" GPU—Acer's pre-launch hype material has outlined an ambitious resolution target: "enjoy stunning 8K visuals and use AI tools to craft your next masterpiece based on RDNA 4 architecture." Acer acts as a board partner for Team Red and Team Blue—surprisingly, the firm's pre-game teaser does not outline an upcoming Intel Xe2 "Battlemage" desktop product line. Instead, a fresh-ish Nitro Arc "Alchemist" A380 LP 6 GB model was placed next to the aforementioned Radeon RX 9060 XT card. So far—during day zero booth inspections—the TechPowerUp has not stumbled upon any physical examples of cheaper RDNA 4 options.

On May 9, the South Korean Radio Agency (RRA) logged four unannounced ASRock Radeon RX 9060 XT SKUs. Since early Spring, harukaze5719 has kept a watchful eye on intriguing pre-launch registrations. For example, they discovered Gigabyte's Gaming OC 16 GB and 8 GB models over a month ago. AMD's board partners seem to be prepping custom options; well in advance of next week's teased unveiling (at Computex 2025). ASRock seems to be cutting things quite fine, with their registering of (allegedly) Navi 44 XT GPU-based Steel Legend and Challenger models. Rumors of a canceled AMD Radeon RX 9060 XT 8 GB (GDDR6) variant emerged a while back, but certain insiders have insisted that this cheaper option will arrive alongside a 16 GB sibling.

Recent RRA filings indicate an upcoming two-pronged approach, involving a good number of AIBs. Buoyed by harukaze5719's fresh findings, VideoCardz carried out additional detective work. They soon unearthed "in stock" ASRock Steel Legend and Challenger factory overclocked Radeon RX 9060 XT cards. Apparently, a Vietnamese store is already courting local client interest—prices are not visible on the unnamed shop's webstore, but staffers are reportedly "encouraging customers to inquire" about costs of ownership. Gaming GPU enthusiasts will recall the amusingly extra early arrival of custom Radeon RX 9070 XT and RX 9070 retail stock in Europe; in anticipation of a supposedly scuppered January launch window. Unlike its main rival, AMD seems to be running on schedule with its next wave of RDNA 4 gaming products.

In an almost uncanny case of recent history repeating itself, AMD has kicked off another RDNA 4 new product teaser campaign. Today's reminder—regarding Jack Huynh's upcoming Computex presentation—included a promotional render of a stubbier dual-fan Radeon RX 9000 Series card design. Casting our memories back to late January (2025), Team Red rolled out an ill-timed advertisement—not long after the official delay of Radeon RX 9070 XT and non-XT cards. Despite denying the existence of "Made-By-AMD" (MBA) reference designs, AMD staffers were likely enraged by Chinese black market channel offerings of alleged "real deal examples." Days after first wave RDNA 4's March 6 global launch, a triple-fan specimen was outed.

Throughout early Q2, insiders and members of the Chiphell forum have played around with Radeon RX 9070 XT and RX 9070 MBA cards. The latter unit (see photo below) seems to utilize a dual-fan configuration in a fairly long enclosure format. Team Red's latest promo post likely points to a forthcoming unveiling of Radeon RX 9060 XT partner models, but curious industry observers will be wondering whether the shorter reference design actually exists in real life. As per usual, a tiny disclaimer claims otherwise: "Artistic Render. Not Available For Purchase." So far, leaks have suggested the presence of Acer, ASUS, GIGABYTE, and XFX custom options during introductory proceedings. Today's refresher outlined upcoming new product categories and partner contributions: "join AMD on May 21 as we reveal what's next in gaming, AI PCs, and more. (Our) SVP and GM of Computing and Graphics Group, along with industry leaders and partners showcase what is built to power the next level."

MSI's European spare parts website has listed a rear plate that can fit on its previous and current generation 7-inch handheld gaming PCs. Eagle-eyed investigators noted the presence of a third unannounced option within a compatible device list: "Claw 7 A2HM (MS-1T41)." In addition, observers noticed that the Japan Audio Society had granted certification back in February. These discoveries have fired up online debates once again; regarding a theorized AMD Ryzen Z2 APU-powered portable gaming device. An early 2025 leak suggested MSI's preparing of an unusual Claw model; apparently breaking away from their "traditional" selection of Intel SoCs. Last month, Haze2K1 reiterated their belief that the manufacturer would eventually release a new handheld with Ryzen Z2 Extreme silicon onboard. Given the complete lack of a "AI" tags/branding, MSI's speculated "Claw A2HM" SKU could arrive with a less potent NPU (or completely disabled). Current-gen marketing campaigns have hyped up the Claw 8 AI+ and Claw 7 AI+'s Copilot+ credentials; driven by Team Blue's "Lunar Lake" Core Ultra 7 258V all-in-one processor.

Very recently, the Taiwanese tech firm introduced a "Polar Tempest Edition" 8-inch flagship spin-off—industry experts posit that MSI is attempting to elevate the Claw 8 AI+'s profile, following a reportedly stuttery early 2025 global rollout. Closer analysis of the "officially" leaked "A2HM" identifier points to a possible "Arrow Lake-H" visitation. As outlined by VideoCardz, this handheld could be based around some sort of Intel Core Ultra 200H chip. Elaborating on their thought process, the online publication commented: "this platform does not use Xe2 GPU cores but instead features Alchemist+ or Xe-LPG+, which is also said to deliver solid gaming performance thanks to XMX core support. Arrow Lake-H also lacks a powerful NPU, which means it doesn't meet the criteria for Copilot+ PC, so Core 200H makes sense." Three months ago, Team Blue's Robert Hallock revealed "big plans" involving futuristic "Arrow Lake-H" and "Panther Lake" portable gaming products. Given recent-ish certification to retail release timelines, handheld enthusiasts could be sizing up MSI's "Claw 7 A2HM" by late summer.

NVIDIA today received multiple accolades at COMPUTEX's Best Choice Awards, in recognition of innovation across the company. The NVIDIA GeForce RTX 5090 GPU won the Gaming and Entertainment category award; the NVIDIA Quantum-X Photonics InfiniBand switch system won the Networking and Communication category award; NVIDIA DGX Spark won the Computer and System category award; and the NVIDIA GB200 NVL72 system and NVIDIA Cosmos world foundation model development platform won Golden Awards. The awards recognize the outstanding functionality, innovation and market promise of technologies in each category. Jensen Huang, founder and CEO of NVIDIA, will deliver a keynote at COMPUTEX on Monday, May 19, at 11 a.m. Taiwan time.

GB200 NVL72 and NVIDIA Cosmos Go Gold
NVIDIA GB200 NVL72 and NVIDIA Cosmos each won Golden Awards. The NVIDIA GB200 NVL72 system connects 36 NVIDIA Grace CPUs and 72 NVIDIA Blackwell GPUs in a rack-scale design. It delivers 1.4 exaflops of AI performance and 30 terabytes of fast memory, as well as 30x faster real-time trillion-parameter large language model inference with 25x energy efficiency compared with the NVIDIA H100 GPU. By design, the GB200 NVL72 accelerates the most compute-intensive AI and high-performance computing workloads, including AI training and data processing for engineering design and simulation. NVIDIA Cosmos accelerates physical AI development by enabling developers to build and deploy world foundation models with unprecedented speed and scale.

MSI IPC, a global leader in industrial computing and AI-driven solutions, is set to unveil its latest innovations at COMPUTEX 2025, held from May 20 to 23 at the Taipei Nangang Exhibition Center. Visitors can explore MSI IPC's cutting-edge technologies at Booth J0506, Hall 1, 1F.

Introducing the EdgeXpert MS-C931: A Desktop AI Supercomputer
MSI IPC will unveil the EdgeXpert MS-C931, a desktop AI supercomputer built on the NVIDIA DGX Spark platform. Powered by the NVIDIA GB10 Grace Blackwell Superchip, the EdgeXpert MS-C931 delivers 1,000 AI TOPS FP4 performance, equipped with high-speed ConnectX 7 networking, 128 GB unified memory, and support for large language models. Designed for AI developers and researchers, it is ideal for applications in education, finance, and healthcare industries.

GEEEK, a Taiwanese SFF chassis brand, announced the Rhino open-air frame cases series designed for compact PC builds, featuring a tilted design with acrylic panels and aluminium block construction. The GEEEK Rhino M microATX measures 305 × 250 × 275 mm (excluding protrusions) and supports microATX and Mini-ITX motherboards. This model offers unrestricted compatibility with all standard air and liquid CPU coolers and graphics cards of any length. The open design provides complete freedom for cooling solutions, supporting radiator sizes ranging from 120 mm to 360 mm. Power supply compatibility includes ATX, SFX, and SFX-L formats with no length limitations.

The smaller Rhino S case is designed exclusively for Mini-ITX motherboards, featuring a more compact footprint of 300 × 200 × 250 mm (excluding protrusions) and weighs 1.1 kg. Despite its smaller size, it maintains the same unrestricted compatibility for CPU coolers and graphics cards as Rhino M. The case supports radiators up to 360 mm and accommodates ATX, SFX, and SFX-L power supplies, however, with a maximum length of 140 mm (maximum 150 mm). The white version features slightly thicker acrylic panels, which may make it a bit more difficult to secure the graphics card, according to the product page.

The Hardware Unboxed team has unleashed some of its sarcastic Aussie wit; in response to an alleged manipulation of GeForce RTX 5060 (non-Ti) review day conditions. In an online dig—directed at Team Green leadership—the Australian media outlet's social media account parodied NVIDIA new product decision-making: "we're not hiding the RTX 5060, we're very proud of it and gamers will love it. Also, we're going to launch the RTX 5060 on May 19th during Computex, and although reviewers have cards right now, we won't be releasing the driver until they go on sale." Mid-way through April, Hardware Unboxed's Tim Schiesser voiced his displeasure regarding a complete lack of GeForce RTX 5060 Ti 8 GB evaluation samples—only 16 GB variants were sent out to testers. Curious professional reviewers opted into buying these cheaper variants (out of pocket), including TechPowerUp's W1zzard. Our head honcho's reckoning—of a custom Gainward effort—pointed out far too many compromises.

In a follow-up post, Hardware Unboxed's social media rep took a more measured approach with their disapproval of "controlled conditions." Clarifying the "context" of their earlier rant, they explained: "NVIDIA are trying to hide the RTX 5060, just as they did the RTX 5060 Ti 8 GB. The strategy here is to release it the week of Computex when most of the tech media are in Taiwan attending the show. They're also blocking reviewers from accessing the driver early to evaluate the RTX 5060 and provide reviews at the time of release. So as it stands I have multiple RTX 5060 samples, and I won't be able to review any of them until about a week after they go on sale." VideoCardz, and other critics/watchers believe that a rumored "rushed" development of GeForce RTX 5060-series cards (Ti and non-Ti) resulted in an uninspiring repeat rollout of 8 GB and 16 GB VRAM configurations—albeit upgraded to GDDR7 standards.

Earlier in the week, reports indicated the potential introduction of an Intel Xe2 "Battlemage" B770 gaming graphics card at Computex 2025. Last night, a Team Blue tweet confirmed forthcoming product unveilings: "new Intel Arc Pro GPUs are on the way. See you in Taipei!" In the months leading up to this important trade event, industry watchdogs have drummed up speculation about "Battlemage's" future (or fate). Whispers of 24 GB VRAM-equipped variants emerged late last year—around late January, these theories were connected to an official leak: "3 new PCI IDs for BMG."

Unsurprisingly, VideoCardz has weighed in with some new inside track info—they propose that one of Intel's upcoming professional options will be an "Arc Pro B60 24 GB" model, aka "Developer Edition" (an alleged in-house reference). Despite Sparkle HQ downplaying recent "rogue claims," a company rep (in China) alluded to a possible May/June release of their own custom 24 GB "Battlemage" productivity-oriented card. VideoCardz has picked up on rumors, regarding the "Arc Pro B60's" internal setup. They propose Team Blue's selection of the familiar "BMG-G21" GPU; as used by their Arc Xe2 B580 12 GB and B570 10 GB designs. According to an unnamed inside source, this professional/workstation variant will stick with the usual 192-bit memory interface. Intel's Computex 2025 new product teaser provided a big clue about the speculated "B60" model's cooling solution.

Join us at Cooler Master HQ for an exclusive COMPUTEX experience. From cooling the demands of AI and high-performance systems to enabling everyday reliability and modular design, discover how we're shaping the future through smarter, more adaptable solutions. Cooler Master invites you to experience the future of tech—live from our headquarters. As the tech landscape evolves, we're pushing cooling further: powering the rise of AI, delivering smarter performance for everyday systems, and unlocking new possibilities through modular design. This year's showcase brings it all together under one vision.

One Cooler Master—Where Cooling Drives the Future
It's the fusion of our industrial and consumer expertise. For over 30 years, we've built advanced cooling solutions for businesses, including: AI data centers, telecom, and mobile—then brought that same innovation to gamers, creators, and everyday users. One Cooler Master reflects our unified vision: engineering that drives performance at every level.

Around late April, MediaTek confirmed that their CEO—Dr. Rick Tsai—will be delivering a big keynote speech—on May 20—at this month's Computex 2025 trade show. The company's preamble focuses on their "driving of AI innovation—from edge to cloud," but industry moles propose a surprise new product introduction during proceedings. MediaTek and NVIDIA have collaborated on a number of projects; the most visible being automative solutions. Late last year, intriguing Arm-based rumors emerged online—with Team Green allegedly working on a first time attempt at breaking into the high-end CPU consumer market segment; perhaps with the leveraging of "Blackwell" GPU architecture. MediaTek was reportedly placed in the equation, due to expertise accumulated from their devising of modern Dimensity "big core" mobile processor designs. At the start of 2025, data miners presented evidence of Lenovo seeking new engineering talent. Their job description mentioned a mysterious NVIDIA "N1x" SoC.

Further conjecture painted a fanciful picture of forthcoming "high-end N1x and mid-tier N1 (non-X)" models—with potential flagship devices launching later on this year. According to ComputerBase.de, an unannounced "GB10" PC chip could be the result of NVIDIA and MediaTek's rumored "AI PC" joint venture. Yesterday's news article divulged: "currently (this) product (can be) found in NVIDIA DGX Spark (platforms), and similarly equipped partner solutions. The systems, available starting at $3000, are aimed at AI developers who can test LLMs locally before moving them to the data center. The chip combines a 'Blackwell' GPU with a 'Grace' Arm CPU (in order) to create an SoC with 128 GB LPDDR5X, and a 1 TB or 4 TB SSD. The 'GB10' offers a GPU with one petaflop of FP4 performance (with sparsity)." ComputerBase reckons that the integrated graphics solution makes use of familiar properties—namely "5th-generation Tensor Cores and 4th-generation RT Cores"—from GeForce RTX 50-series graphics cards. When discussing the design's "Grace CPU" setup, the publication's report outlined a total provision of: "20 Arm cores, including 10 Cortex-X925 and 10 Cortex-A725. The whole thing sits on a board measuring around 150 × 150 mm—for comparison: the classic NUC board format is 104 × 101 mm."